Neuroscheduling for Remote Estimation

TL;DR

This paper introduces neuroscheduling, a data-driven approach for optimal data source selection in distributed systems with real-time constraints, revealing that the optimal scheduler is nonlinear for Gaussian sources.

Contribution

It proposes a novel neuroscheduling methodology using linear function approximation to find optimal scheduling strategies in complex remote estimation scenarios.

Findings

Optimal scheduler and estimator are nonlinear for Gaussian sources.

Necessary and sufficient conditions for avoiding overfitting in neuroscheduling.

Neuroscheduling outperforms traditional linear approaches.

Abstract

Many modern distributed systems consist of devices that generate more data than what can be transmitted via a communication link in near real time with high-fidelity. We consider the scheduling problem in which a device has access to multiple data sources, but at any moment, only one of them is revealed in real-time to a remote receiver. Even when the sources are Gaussian, and the fidelity criterion is the mean squared error, the globally optimal data selection strategy is not known. We propose a data-driven methodology to search for the elusive optimal solution using linear function approximation approach called neuroscheduling and establish necessary and sufficient conditions for the optimal scheduler to not over fit training data. Additionally, we present several numerical results that show that the globally optimal scheduler and estimator pair to the Gaussian case are nonlinear.